Feghali et al reported the R2eD AVM score to evaluate a patient with a cerebral artertiovenous malformation (AVM). This can help to identify an AVM at risk for rupture with hemorrhage. The authors are from the Johns Hopkins University.
Patient selection: cerebral arteriovenous malformation
Parameters:
(1) race (Asian, Black, Other, White)
(2) location
(3) size
(4) venous drainage
(5) number of feeding arteries
|
Parameter
|
Finding
|
Beta Coefficients
|
Simple Score
|
|
race
|
White
|
0
|
0
|
|
|
other (non-White)
|
0.912
|
2
|
|
location
|
deep (thalamus, basal ganglia, brain stem, cerebellum)
|
0.516
|
1
|
|
|
not deep
|
0
|
0
|
|
size
|
small (< 3 cm)
|
0.486
|
1
|
|
|
large (>= 3 cm)
|
0
|
0
|
|
venous drainage
|
exclusively deep
|
0.547
|
1
|
|
|
other
|
0
|
0
|
|
number of feeding arteries
|
1
|
0.456
|
1
|
|
|
2 or more
|
0
|
0
|
X =
= SUM(points for all 5 parameters) - 1.676
probability of rupture with hemorrhage at presentation=
= 1 / (1 + EXP((-1) * X))
simple score =
= SUM(points for the 5 parameters)
Interpretation:
• minimum simple score: 0
• maximum simple score: 6
• The higher the score the greater the risk of hemorrhage.
|
Simple Score
|
Probability of Hemorrhage
|
|
0
|
16%
|
|
1
|
23%
|
|
2
|
32%
|
|
3
|
45%
|
|
4
|
54%
|
|
5
|
67%
|
|
6
|
78%
|
Performance:
• The area under the ROC curve is around 0.70.
